Notes on Features. Ling 571 Deep Techniques for NLP February 10, 2014

Transcription

1 Notes on Features Ling 571 Deep Techniques for NLP February 10, 2014

2 Feature Grammar in NLTK NLTK supports feature-based grammars Includes ways of associating features with CFG rules Includes readers for feature grammars.fcfg files Includes parsers Nltk.parse.FeatureEarleyChartParser

3 Feature Structures >>> fs1 = nltk.featstruct( [NUM= pl ] ) >>> print fs1 [NUM= pl ] >>> print fs1[ NUM ] pl More complex structure >>> fs2 = nltk.featstruct( [POS= N, AGR=[NUM= pl,per=3]] )

4 Reentrant Feature Structures First instance Parenthesized integer: (1) Subsequent instances: Pointer : -> (1) >>> print nltk.featstruct("[a='a', B=(1)[C='c'], D->(1)] [ A = a ] [ B = (1) [ C = c ]] [ D -> (1) ]

5 Augmenting Grammars Attach feature information to non-terminals, on N[AGR=[NUM='pl']] -> 'students N[AGR=[NUM= sg']] -> 'student So far, all values are literal or reentrant Variables allow generalization:?a Allows underspecification, e.g. Det[GEN=?a] NP[AGR=?a] -> Det[AGR=?a] N[AGR=?a]

6 Mechanics >>> fs3 = nltk.featstruct(num= pl,per=3) >>> fs4 = nltk.featstruct(num= pl ) >>> print fs4.unify(fs3) [NUM = pl ] [PER = 3 ]

7 Morphosyntactic Features Grammatical feature that influences morphological or syntactic behavior English: Number: Dog, dogs Person: Am; are; is Case: I me; he him; etc Countability:

8 Semantic Features Grammatical features that influence semantic(meaning) behavior of associated units E.g.:

9 Semantic Features Grammatical features that influence semantic(meaning) behavior of associated units E.g.:?The rocks slept.

10 Semantic Features Grammatical features that influence semantic(meaning) behavior of associated units E.g.:?The rocks slept.?colorless green ideas sleep furiously.

11 Semantic Features Many proposed: Animacy: +/- Natural gender: masculine, feminine, neuter Human: +/- Adult: +/- Liquid: +/- Etc. The milk spilled.?the cat spilled.

12 Examples The climber hiked for six hours. The climber hiked on Saturday. The climber reached the summit on Saturday. *The climber reached the summit for six hours. Contrast:

13 Examples The climber hiked for six hours. The climber hiked on Saturday. The climber reached the summit on Saturday. *The climber reached the summit for six hours. Contrast: Achievement vs activity

14 Semantic features & Parsing Can filter some classes of ambiguity Old men and women slept. (Old men) and (women) slept. (Old (men and women)) slept. Sleeping people and books lie flat. (Sleeping people) and (books) lie flat. (Sleeping (people and books ))lie flat.

15 Semantic features & Parsing Can filter some classes of ambiguity Old men and women slept. (Old men) and (women) slept. (Old (men and women)) slept. Sleeping people and books lie flat. (Sleeping people) and (books) lie flat. *(Sleeping (people and books ))lie flat.

16 Summary Features Enable compact representation of grammatical constraints Capture basic linguistic patterns Unification Creates and maintains consistency over features Integration with parsing allows filtering of illformed analyses

17 More Complex German Subject singular, masc der Hund The dog Example Subject plural, masc die Hunde The dogs

18 More Complex German Example Objects determined by verb Dative singular, masc dem Hund The dog Accusative plural, masc die Hunde The dogs

19 Contrast Subject: Die Katze The cat Subject: plural Die Katzen The cats

20 Contrast Object: Die Katze The cat Object: Der Katze The cat

21 Analysis What are the key contrasts? Number Singular, plural Gender Masc, Fem,. Case: Subject (nom), dative, accusative,. + Interactions

22 Feature Interaction Interactions of German case, number, gender Case Masc Fem Neut PL Nom Der Die Das Die Gen Des Der Des Den Dat Dem Der Dem Den Acc Den Die Das Die

23 Examples of Interaction Die The.Nom.Fem.sg The cat sees the dog Katze Cat.3.FEM.SG Sieht See.3.sg Den The.Acc.Masc.sg Hund Dog.3.Masc.sg

24 Examples of Interaction Die The.Nom.Fem.sg The cat sees the dog Katze Cat.3.FEM.SG Sieht See.3.sg Den The.Acc.Masc.sg Hund Dog.3.Masc.sg *Die The.Nom.Fem.sg The cat sees the dog Katze Cat.3.FEM.SG Sieht See.3.sg Dem The.Dat.Masc.sg Hund Dog.3.Masc.sg

25 Examples of Interaction Die The.Nom.Fem.sg The cat sees the dog Katze Cat.3.FEM.SG Sieht See.3.sg Den The.Acc.Masc.sg Hund Dog.3.Masc.sg *Die The.Nom.Fem.sg The cat sees the dog Katze Cat.3.FEM.SG Sieht See.3.sg Dem The.Dat.Masc.sg Hund Dog.3.Masc.sg Die The.Nom.Fem.sg The cat helps the dog Katze Cat.3.FEM.SG hilft help.3.sg Dem The.Dat.Masc.sg Hund Dog.3.Masc.sg

26 Examples of Interaction Die The.Nom.Fem.sg The cat sees the dog Katze Cat.3.FEM.SG Sieht See.3.sg Den The.Acc.Masc.sg Hund Dog.3.Masc.sg *Die The.Nom.Fem.sg The cat sees the dog Katze Cat.3.FEM.SG Sieht See.3.sg Dem The.Dat.Masc.sg Hund Dog.3.Masc.sg Die The.Nom.Fem.sg The cat helps the dog Katze Cat.3.FEM.SG hilft help.3.sg Dem The.Dat.Masc.sg Hund Dog.3.Masc.sg *Die The.Nom.Fem.sg The cat sees the dog Katze Cat.3.FEM.SG hilft help.3.sg Dem The.Acc.Masc.sg Hund Dog.3.Masc.sg German verbs in, at least, 2 classes: assign diff t object case